pandas学习与使用2

继续学习pandas库，上一节主要介绍了Series，这一节主要是DataFrame结构的用法。运行环境python2.7

#!usr/bin/python3
# coding:utf-8
# pandas 使用DataFrame

import numpy as np
import pandas as pd

dates = pd.date_range('20171231', periods=6)
print "dates:\n", dates

#                    a         b         c         d
# 2017-12-31  0.544078  0.389521  0.097052  0.942329
# 2018-01-01  0.474514  0.456605  0.750682  0.683513
# 2018-01-02  0.537973  0.230534  0.216569  0.015208
# 2018-01-03  0.320855  0.295421  0.342874  0.808681
# 2018-01-04  0.649339  0.678842  0.390282  0.692622
# 2018-01-05  0.041877  0.197155  0.384499  0.301309
# index 为纵向坐标的，column为横向坐标

df = pd.DataFrame(np.random.rand(6, 4), index=dates, columns=['a', 'b', 'c', 'd'])
print "df:\n", df
print "dtypes:\n", df.dtypes
print "columns:\n", df.columns
print "values:\n", df.values
# describe是DataFrame的详细信息，包括count、mena、std等值
print "describe:\n", df.describe()

#               a         b         c         d
# count  6.000000  6.000000  6.000000  6.000000
# mean   0.428106  0.374680  0.363660  0.573944
# std    0.217850  0.177776  0.220942  0.347394
# min    0.041877  0.197155  0.097052  0.015208
# 25%    0.359270  0.246756  0.248145  0.396860
# 50%    0.506243  0.342471  0.363687  0.688067
# 75%    0.542552  0.439834  0.388836  0.779666
# max    0.649339  0.678842  0.750682  0.942329
print "转置：\n", df.T

# 按照索引排序 ascending=False 倒序列
print "按照索引排序:\n", df.sort_index(axis=1, ascending=False)
# 按照值排序
print "按照值排序:\n", df.sort_values(by='a')
print "______________________"

df1 = pd.DataFrame(np.arange(12).reshape((3, 4)))
print "df1:\n", df1
print "df1.dtypes:\n", df1.dtypes

df2 = pd.DataFrame({'A': 1, 'B': pd.Timestamp('20180101'), 'C': pd.Series(1, index=list(range(4)), dtype=float)})
print "df2:\n", df2

# DataFrame的构造
# pd.DataFrame() 参数： 1、二维array； 2、Series 列表； 3、value为Series的字典；

# 1、二维array
s1 = np.array([1, 2, 3, 4])
s2 = np.array([5, 6, 7, 8])
dataframe1 = pd.DataFrame([s1, s2])
print "dataframe1:\n", dataframe1

# 2、Series 列表
s1 = pd.Series(np.array([1, 2, 3, 4]))
s2 = pd.Series(np.array([5, 6, 7, 8]))
dataframe2 = pd.DataFrame([s1, s2])
print "dataframe2:\n", dataframe2

# 3、value为Series的字典
s1 = pd.Series(np.array([1, 2, 3, 4]))
s2 = pd.Series(np.array([5, 6, 7, 8]))
dataframe3 = pd.DataFrame({"a": s1, "b": s2});
print "dataframe3:\n", dataframe3

# DataFrame的属性
print "dataframe3的columns的值：\n", dataframe3.columns
print "dataframe3的形状：\n", dataframe3.shape
print "dataframe3的index的值：\n", list(dataframe3.index)
print "dataframe3的value的值：\n", dataframe3.values

# DataFrame的if-then操作
# df.ix[条件，then操作区域]
df = pd.DataFrame({"A": [1, 2, 3, 4], "B": [5, 6, 7, 8], "C": [1, 1, 1, 1]})
print "修改前：\n", df
# if(df.A > 2),"B"=-1
df.ix[df.A > 2, 'B'] = -1
print "修改后：\n", df

# 使用numpy.where
df = pd.DataFrame({"A": [1, 2, 3, 4], "B": [5, 6, 7, 8], "C": [1, 1, 1, 1]})
# np.where(条件，then，else)
df["then"] = np.where(df.A < 3, 1, 0)
print "修改后：\n", df

# 直接取值df.[]
df = pd.DataFrame({"A": [1, 2, 3, 4], "B": [5, 6, 7, 8], "C": [1, 1, 1, 1]})
df = df[df.A >= 2]
print "df[df.A >= 2]:\n", df

# 使用.loc[]
df = pd.DataFrame({"A": [1, 2, 3, 4], "B": [5, 6, 7, 8], "C": [1, 1, 1, 1]})
df = df.loc[df.A > 2]
print "df.loc[df.A > 2]:\n", df

df = pd.DataFrame({'animal': 'cat dog cat fish dog cat cat'.split(),
                   'size': list('SSMMMLL'),
                   'weight': [8, 10, 11, 1, 20, 12, 12],
                   'adult': [False] * 5 + [True] * 2})
print "df:\n", df

以下是上述代码的运行结果

D:\software\Anaconda2\python.exe D:/PycharmProjects/Learn/learn_panda/learn_pd2.py
dates:
DatetimeIndex(['2017-12-31', '2018-01-01', '2018-01-02', '2018-01-03',
               '2018-01-04', '2018-01-05'],
              dtype='datetime64[ns]', freq='D')
df:
                   a         b         c         d
2017-12-31  0.138818  0.723637  0.095903  0.368734
2018-01-01  0.887222  0.371972  0.269801  0.250211
2018-01-02  0.121923  0.861907  0.955060  0.770644
2018-01-03  0.886326  0.606760  0.418799  0.745113
2018-01-04  0.181974  0.611008  0.144484  0.385850
2018-01-05  0.422556  0.294514  0.143726  0.420354
dtypes:
a    float64
b    float64
c    float64
d    float64
dtype: object
columns:
Index([u'a', u'b', u'c', u'd'], dtype='object')
values:
[[0.13881833 0.72363691 0.09590252 0.36873399]
 [0.88722168 0.37197237 0.26980146 0.25021065]
 [0.12192265 0.86190715 0.95506014 0.77064439]
 [0.88632573 0.6067601  0.41879897 0.74511262]
 [0.1819737  0.61100784 0.14448442 0.38585008]
 [0.42255628 0.29451366 0.1437259  0.42035442]]
describe:
              a         b         c         d
count  6.000000  6.000000  6.000000  6.000000
mean   0.439803  0.578300  0.337962  0.490151
std    0.362761  0.212857  0.324267  0.215288
min    0.121923  0.294514  0.095903  0.250211
25%    0.149607  0.430669  0.143916  0.373013
50%    0.302265  0.608884  0.207143  0.403102
75%    0.770383  0.695480  0.381550  0.663923
max    0.887222  0.861907  0.955060  0.770644
转置：
   2017-12-31  2018-01-01     ...      2018-01-04  2018-01-05
a    0.138818    0.887222     ...        0.181974    0.422556
b    0.723637    0.371972     ...        0.611008    0.294514
c    0.095903    0.269801     ...        0.144484    0.143726
d    0.368734    0.250211     ...        0.385850    0.420354

[4 rows x 6 columns]
按照索引排序:
                   d         c         b         a
2017-12-31  0.368734  0.095903  0.723637  0.138818
2018-01-01  0.250211  0.269801  0.371972  0.887222
2018-01-02  0.770644  0.955060  0.861907  0.121923
2018-01-03  0.745113  0.418799  0.606760  0.886326
2018-01-04  0.385850  0.144484  0.611008  0.181974
2018-01-05  0.420354  0.143726  0.294514  0.422556
按照值排序:
                   a         b         c         d
2018-01-02  0.121923  0.861907  0.955060  0.770644
2017-12-31  0.138818  0.723637  0.095903  0.368734
2018-01-04  0.181974  0.611008  0.144484  0.385850
2018-01-05  0.422556  0.294514  0.143726  0.420354
2018-01-03  0.886326  0.606760  0.418799  0.745113
2018-01-01  0.887222  0.371972  0.269801  0.250211
______________________
df1:
   0  1   2   3
0  0  1   2   3
1  4  5   6   7
2  8  9  10  11
df1.dtypes:
0    int32
1    int32
2    int32
3    int32
dtype: object
df2:
   A          B    C
0  1 2018-01-01  1.0
1  1 2018-01-01  1.0
2  1 2018-01-01  1.0
3  1 2018-01-01  1.0
dataframe1:
   0  1  2  3
0  1  2  3  4
1  5  6  7  8
dataframe2:
   0  1  2  3
0  1  2  3  4
1  5  6  7  8
dataframe3:
   a  b
0  1  5
1  2  6
2  3  7
3  4  8
dataframe3的columns的值：
Index([u'a', u'b'], dtype='object')
dataframe3的形状：
(4, 2)
dataframe3的index的值：
[0, 1, 2, 3]
dataframe3的value的值：
[[1 5]
 [2 6]
 [3 7]
 [4 8]]
修改前：
   A  B  C
0  1  5  1
1  2  6  1
2  3  7  1
3  4  8  1
修改后：
   A  B  C
0  1  5  1
1  2  6  1
2  3 -1  1
3  4 -1  1
修改后：
   A  B  C  then
0  1  5  1     1
1  2  6  1     1
2  3  7  1     0
3  4  8  1     0
df[df.A >= 2]:
   A  B  C
1  2  6  1
2  3  7  1
3  4  8  1
df.loc[df.A > 2]:
   A  B  C
2  3  7  1
3  4  8  1
df:
   adult animal size  weight
0  False    cat    S       8
1  False    dog    S      10
2  False    cat    M      11
3  False   fish    M       1
4  False    dog    M      20
5   True    cat    L      12
6   True    cat    L      12

Process finished with exit code 0